aTENNuate：在原始音频上基于深度SSM的优化实时语音增强
aTENNuate: Optimized Real-time Speech Enhancement with Deep SSMs on Raw Audio
AI总结：
本研究提出深度状态空间自编码器aTENNuate，用于端到端在线原始语音增强，在多个基准测试中于PESQ、参数量等指标上优于现有实时去噪模型，低资源环境下仍表现优异。
摘要：
我们提出了aTENNuate，这是一种简单的深度状态空间自编码器，专为高效的端到端在线原始语音增强任务配置。该网络的性能主要在原始语音去噪任务上评估，同时还在超分辨率、去量化等任务上进行了额外测试。我们在VoiceBank + DEMAND和Microsoft DNS1合成测试集上对aTENNuate开展基准测试。该网络在PESQ分数、参数量、MACs和延迟方面均优于此前的实时去噪模型。即便作为原始波形处理模型，它仍能保持与干净信号的高保真度，可听伪影极少。此外，当含噪输入被压缩至4000Hz、4比特时，该模型依然性能优异，表明其在低资源环境下具备通用语音增强能力。可通过pip install attenuate试用。
英文摘要：
We present aTENNuate, a simple deep state-space autoencoder configured for efficient online raw speech enhancement in an end-to-end fashion. The network's performance is primarily evaluated on raw speech denoising, with additional assessments on tasks such as super-resolution and de-quantization. We benchmark aTENNuate on the VoiceBank + DEMAND and the Microsoft DNS1 synthetic test sets. The network outperforms previous real-time denoising models in terms of PESQ score, parameter count, MACs, and latency. Even as a raw waveform processing model, the model maintains high fidelity to the clean signal with minimal audible artifacts. In addition, the model remains performant even when the noisy input is compressed down to 4000Hz and 4 bits, suggesting general speech enhancement capabilities in low-resource environments. Try it out by pip install attenuate
机构、总结、中文摘要均由AI分析而来，可能存在错误，请以论文正文为准。
